Abstract
Delay and Disruption Tolerant networks (DTNs) are an alternative means of providing Internet access to remote communities, where there is a lack of a telecommunications infrastructure. Communication in this scenario is carried out through opportunistic encounters between nodes, called contacts, in which messages are forwarded to the destination. Predicting information about a contact, at the time of its occurrence, allows strategies to be adopted for optimizing message dispatch to improve the delivery rate and reduce the delivery delay in DTNs. This paper outlines a model to predict the amount of data that can be transferred, during a DTN encounter, based on real experiments conducted in the Amazon region. The results show that the proposed model has the advantage of including environmental conditions, such as soil and vegetation in the Contact Volume prediction, plus parameters like Radio Frequency power, the wireless networking standard, and the type and height of the radio antennas.